Long-term projections of global food security with R&D-driven technological progress

Food security is one of the largest challenges facing mankind in the next half century (as acknowledged for instance by UNDP, 2012 and UNEP, 2012). The projections of population growth warn that by 2050 the agricultural sector will have to feed 9 billion people, which requires doubling the current levels of food production. Due to the constrained expansion of agricultural land, technical progress is required that will drive agricultural productivity and therefore make an important contribution to future improvements of global food security. Long-term projections of global food security should take into account that food security is a multidimensional concept, which includes dimensions of availability, accessibility, utilization, and stability (FAO, 2000). The availability dimension is associated with the physical supply of food which increasingly relies on technical change in agriculture, triggered by investments to research and development (Avila and Evenson, 2010). The most contemporary global state-of-the art CGE models however do not accurately reflect this linkage (i.e. models included in the Agmip Global Economic Comparison Project). The accessibility dimension is related to households’ income and the evolution and variability of food prices (Sen, 1981). However, the experiment performed by Robison et al. (2013) showed that food prices projections may be highly diverging under various technical change assumptions that specify the type of factor-bias and the inter-sectoral spillovers. The paper aims at providing projections of food security in a GTAP-based CGE model MAGNET (Modular Applied General Equilibrium Tool), developed at LEI. Main contributions of this paper are threefold: i) it provides long-term projections of food security with R&D driven endogenous technical change and thus it partially opens the “black-box” in modelling technical change, ii) it builds on empirical estimates of endogenous technical change and thus it increases the reliability of food price projections, iii) it encompasses various dimensions of food security in a general equilibrium framework, which enables to capture important inter-sectoral linkages in factor markets and the effects of R&D spillovers across all regions of the world. The modelling approach is based on the incorporation of a specific R&D module into MAGNET, in which two types of R&D activities are distinguished - public and private agricultural R&D. Public agricultural R&D is considered as a land-augmenting research activity, represented mainly by investments into new crop varieties, which is fully demanded by government. Certain lag is considered before R&D is fully transmitted into higher land productivity. Moreover, public R&D spillovers are included in the model based on the regional distance from the global technology frontier. On the other hand, private R&D investments are considered as by-product activities of agricultural input sectors (agriculture, chemical and transport industry) assuming that the R&D investment incentives go hand in hand with the industrial performance. Private R&D investments stimulate land and labor-augmenting technical change via knowledge which acts as a new production factor in the economy. The projections of Food security are obtained for the period 2007 – 2050. There are four baseline scenarios that have been constructed in a joint stakeholder/modelling process and that reflect the future of global economy taking into account dimensions of inequality and sustainability. In this paper, the projections of food security are analysed for these four baseline scenarios assuming that governmental expenditures on R&D follow regional GDP growth. Besides standard indicators such as agricultural production and prices, specific food security indicators including nutritional status of the households are reported.